Adherence and persistence to self-administered disease-modifying therapies in patients with multiple sclerosis: A multisite analysis.

Background: Though there are several disease-modifying therapy (DMT) options for patients with multiple sclerosis (MS), treatment outcomes rely on patient adherence and persistence. Previous studies have demonstrated suboptimal adherence rates and high rates of early treatment discontinuation. Health-system specialty pharmacies (HSPPs) are a growing practice model that have demonstrated adherence and persistence benefits through single site evaluations. Research is needed across multiple HSSPs to understand and validate the outcomes of this practice model.

Methods: A multisite prospective cohort study was performed including patients with at least three fills of a DMT between January 2020 and June 2021 at an HSSP. Patients were excluded due to pregnancy or death. Enrollment occurred for 6 months followed by 12 months of follow-up. Adherence was measured using pharmacy claims to calculate proportion of days covered (PDC) during the follow-up period. Time to non-persistence was calculated as the time from an index DMT fill to the first date of a gap of >60 days between medication exhaust and fulfillment dates. Adherence and persistence calculations were assessed at the therapeutic class level (any self-administered DMT dispensed by the HSSPs). The Kaplan-Meier method was used to present the probability of being persistent, and Cox proportional hazards regression analysis was used to estimate hazard ratios of factors associated with non-persistence, which included age, sex, study site, insurance type, and whether the patient switched medication as potential factors.

Results: The most common self-administered DMTs filled among 968 patients were glatiramer acetate (32%), fingolimod (18%), and dimethyl fumarate (18%). Most patients (96%) did not switch DMT during the study period. The median PDC was 0.97 (interquartile range 0.90-0.99), which was similar across all sites. Patients who had at least one DMT switch were 76% less likely to have a higher PDC than those who did not have any switch after adjusting for other covariates (Odds ratio: 0.24, 95% confidence interval [CI]: 0.14-0.40, p<0.001). Most patients (86%) were persistent to DMT over the 12-month study period. Among those non-persistent, median time to non-persistence was 231 (IQR 177-301) days. Patients who switched medications were 2.4 times more likely to be non-persistent (95% CI: 1.3 - 4.5, p = 0.005). The most common reasons for non-persistence were discontinuation/medication held for an extended period (30%), often due to patient or prescriber decision (75%).

Conclusion: High rates of DMT adherence and persistence were seen among patients serviced by HSSPs, indicating potential benefits of this model for patients with MS. Switching DMTs was associated with lower adherence and persistence and may be an opportunity for added care coordination or resources to optimize therapy transitions.